Dumb question: APO and monochrom

[IkarusJohn]

Advertisement (gone after registration)

7 hours ago, RexGig0 said:

When I was interested in adding a Noctilux 50mm, in early 2022, with the specific goal of letting more light enter my M Type 246 Monochrom camera, rather than upgrade to an M10 Monochrom, I found at least one source that indicated that a particular version left blue light less-well-corrected, which had a specific effect that was/is desirable when shooting black-and-white. I will try to find where I saw/read/heard this.

There was discussion some time ago about the use of red filters on lenses causing focus shift with the Monochrom.  I don’t recall who raised the issue - Lloyd Chambers, perhaps, or Ken Rockwell.  The point was that while a dark red filter will add drama to a Monochrom shot (brightening clouds, darkening skies etc), ir would also exacerbate any lack of apochromatic correction in a lens.

I took my Monochrom out with the Noctilux 0.95 and dark red filter and photographed a boatshed, middle to long(ish) distance wide open and could find no such shift.  Not entirely scientific, but practical. The Noctilux was sufficiently apochromatic for my purposes, in any event.

As to Olaf’s question, there are too many flaws and limitations quoted and repeated which have limited practical impact, in my view (even if they are true).  For example, my v2 35mm Summilux pre-asph is being cleaned in Wetzlar at the moment.  I asked if there was a 6-bit code for it they could mount. They came back to me saying it does not focus well enough for digital, being designed for film.  And yet, it is a favourite of many photographers with digital (see the long thread of images).

The simple answer is, as Jeff states above - a better lens improves the image. Now, how you define “improving” is a debate all of its own. I use a number of APO lenses on my Monochrom, and the increase in fine detail over my long gone M9 is telling.

Quote

… 

do apo lenses have more impact on non- monochrom cameras or not? Is apo relevant for monochrom?

Not in my experience.

By removing the Bayer filter array, the increase in detail with the higher resolving APO designated lenses is more apparent with a Monochrom camera than a non-monochrom camera (for a given MP sensor).  If you read the technical detail on APO lenses, among their other benefits is their increased resolving power measured in lines per millimetre.  That increase in detail is why people make the perhaps unhelpful comment that APO lenses are good for 100MP sensors in full frame format.

If anything, the removal of the Bayer filter array enables a Monochrom camera to reveal more detail in an image than the colour version of the same camera.  Other things may happen, of course, but I can think of no reason in practice why an APO lens should have less impact on a Monochrom camera (whatever “impact” means).

Looking at it another way, if you take, say, an M10-R and take an image with the original 50 Summilux ASPH, then take the same image with the APO 50 Summicron, the image will have “improved”; or at least, should be measurably, technically “better”.  Repeat the process with the M10-M, and the APO image would also be “better”.  Would the impact of the APO lens be greater with the M10-R than with the M10-M?  Who knows?  Probably not, as the M10-M will show more detail than the colour equivalent.  There may be more going on.  Does it matter?

I appreciate this post is longer than it could have been … 🙂

[otto.f]

On 6/30/2023 at 3:05 PM, Olaf_ZG said:

irrelevant

Absolutely not. Especially in green, foliage, the APO is able to render all the different nuances. This is because green is most off focus with a lens that is not corrected for the way all the colors in the spectrum bend to the focus point. If a certain color does not come onto the focus point but a bit farther or closer, it will, also in B&W, result in less sharpness because you only get a part of the spectrum in focus and other part causes blur.

 

Edited July 1, 2023 by otto.f

[otto.f]

14 minutes ago, IkarusJohn said:

I don’t recall who raised the issue - Lloyd Chambers, perhaps, or Ken Rockwell.

After the release of the Leica’s first Monochrome1, LFI magazine immediately showed with examples how orange and red filters have less precise focus, the sharpness lies a bit farther away when using these filters. It’s not that the genius of Ken Rockwell discovered this, it’s just plain optical fysics which is known by Leica.

[Guest]

On 7/1/2023 at 4:42 PM, otto.f said:

After the release of the Leica’s first Monochrome1, LFI magazine immediately showed with examples how orange and red filters have less precise focus, the sharpness lies a bit farther away when using these filters. It’s not that the genius of Ken Rockwell discovered this, it’s just plain optical fysics which is known by Leica.

So (landscape) photographers with monochrome Ms and expensive lenses have the choice between suboptimal sharpness from their dear gear when using “a very good” glass filter in front of the lens

or

overexposed and “cloudless” skies, dresses or flowers in unusual brightness/darkness etc.

??

It could well be that the first sentence (and hence the quote?) is not perfectly correct, because how should a monochrome camera know, that the light reaching the monochrome sensor has an orange cast because of a filter in front of a super-expensive APO lens or because the sun sets / the light is artificial and not daylight?

Edited July 3, 2023 by Apochromat

[IkarusJohn]

3 hours ago, Apochromat said:

So (landscape) photographers with monochrome Ms and expensive lenses have the choice between suboptimal sharpness from their dear gear when using “a very good” glass filter in front of the lens

or

overexposed and “cloudless” skies, dresses or flowers in unusual brightness/darkness etc.

??

It could well be that the first sentence (and hence the quote?) is not perfectly correct, because how should a monochrome camera know, that the light reaching the monochrome sensor has an orange cast because of a filter in front of a super-expensive APO lens or because the sun sets / the light is artificial and not daylight?

There’s more to that story, really.

Start with the rangefinder system - an optic rangefinder mechanism, uncoupled from the lens and sensor; properly adjusted, an image in focus should provide a sharp image, regardless of whether the sensor has a colour filter array or has no such filter.  The monochromatic sensor simply records the light hitting the sensor with no colour filtration. 

Let’s assume, then, that the rangefinder produces a sharp image, regardless of the sensor. No reason not to.  Let’s also assume that the lens is apochromatic - all visible light frequencies hit the sensor at the same point, in focus.  Fair assumption?

Now, I stopped studying physics at university, but as I recall, like a coloured surface which absorbs all visible light except the colour reflected, a filter only transmits one colour. All other parts of the visible spectrum is absorbed by the filter.  So, for a deep red filter, if the red part of the spectrum was hitting the sensor at the same point as all other colours of the spectrum, adding a filter would not “shift” the frequency of the colour red.

That’s basic physics.

It’s not the sensor or the filter that's the issue.  For a properly focused image, if adding a filter introduces back focus, then the lens was not properly apochromatic in the first place.  The filter has zero impact on the rangefinder and the sensor.  What we can conclude is that, at infinity, minor colour shifts, particularly in the red spectrum (at the far end of the colour spectrum) focus may not be as sharp.  If that’s the case, then that lack of sharpness will be present with a monochrome image without the filter or with a colour sensor, taken with the same lens at the same focus.

I like taking landscape images with my Monochrom, and I use filters a lot - I get better images than if I try to emulate the same thing using my M10-D and post process using sliders.  I also tend to use non “APO” lenses like the 75 Summilux and the 50 Summilux ASPH for landscape (for some reason, I don’t think of the Noctilux as a landscape lens, but stopped down I can think of no reason not to).  But, in every case, I stop down and focus carefully. Sharpness is critical to landscapes, in my view.

The simple physics is, adding a filter doesn’t shift wave lengths. They are already there.  If you’re getting out of focus images with filters, your lens was out of focus in the first place.
 

[Jeff S]

Lateral chromatic aberration in the lens produces the red focus shift. More effect with deeper red filters. Same when using film.   

Jeff

Edited July 3, 2023 by Jeff S

[IkarusJohn]

Advertisement (gone after registration)

45 minutes ago, Jeff S said:

Lateral chromatic aberration in the lens produces the red focus shift. More effect with deeper red filters. Same when using film.   

Jeff

Sure, but it was always there. In reality, I’ve found no focus shift using a deep red filter - well corrected lens and depth of field. We are talking about focus shift, not CA.  Even in theory, a well made filter should not cause focus shift.

Edit - Axial (longitudinal) chromatic aberration affects the red spectrum most and occurs across the image.  This is hidden by depth of field, if it’s present.  It affects longer focal lengths more.  Transverse (lateral) chromatic aberration is more prone in wide angle lenses, and affects all parts of the colour spectrum at the edges of the image.

The issue raise by Digilloyd and LFI was the use of a red filter with a Monochrom camera.  Sounds reasonable in theory, but doesn’t stand up, if a properly apochromatic lens is focused properly. If in doubt, depth of field is your friend.

Edited July 3, 2023 by IkarusJohn

[Jeff S]

33 minutes ago, IkarusJohn said:

Sure, but it was always there. In reality, I’ve found no focus shift using a deep red filter - well corrected lens and depth of field. We are talking about focus shift, not CA.  Even in theory, a well made filter should not cause focus shift.

Edit - Axial (longitudinal) chromatic aberration affects the red spectrum most and occurs across the image.  This is hidden by depth of field, if it’s present.  It affects longer focal lengths more.  Transverse (lateral) chromatic aberration is more prone in wide angle lenses, and affects all parts of the colour spectrum at the edges of the image.

The issue raise by Digilloyd and LFI was the use of a red filter with a Monochrom camera.  Sounds reasonable in theory, but doesn’t stand up, if a properly apochromatic lens is focused properly. If in doubt, depth of field is your friend.

We’ve covered this.  Lloyd had the effect, but didn’t understand the cause (Post #16 from Jaap, which you ‘liked’.)

Jeff

Edited July 3, 2023 by Jeff S

[Guest]

1 hour ago, IkarusJohn said:

… In reality, I’ve found…

Edit - Axial (longitudinal) chromatic aberration affects the red spectrum most and occurs across the image.  This is hidden by depth of field, if it’s present.  It affects longer focal lengths more.  Transverse (lateral) chromatic aberration is more prone in wide angle lenses, and affects all parts of the colour spectrum at the edges of the image.

The issue raise by Digilloyd and LFI was the use of a red filter with a Monochrom camera.  Sounds reasonable in theory, but doesn’t stand up, if a properly apochromatic lens is focused properly. If in doubt, depth of field is your friend.

“In reality, I’ve found” is imo more reasonable than musing about articles written by…

Followed by: Edit… a clear explanation.

[IkarusJohn]

1 hour ago, Jeff S said:

We’ve covered this.  Lloyd had the effect, but didn’t understand the cause (Post #16 from Jaap, which you ‘liked’.)

Jeff

Yes, that’s right.

The fact it has been discussed before, doesn’t stop the issue being raised again.  Your exemplary searching skills do help to answer these questions.

In the thread you referenced, I suggested apochromatic lenses largely deal with the issue; in practice.  The post of @jaapv’s I liked was as follows:

Quote

The problem was Digiloyd's ignorance of the cause. It happens on film as well and has been known for as long as rangefinders exist. It has nothing to do with the filter stack. If you have a lens with lateral CA the focus will shift to the red fringe. 

It  becomes more serious as the red deepens. With IR filters it can be severe.

He’s not quite right in his reference to lateral or transverse CA, which tends to affect all colours and results in smearing and colour shift in the corners and around the edges. There’s lots of discussion about “bokeh balls” and butterfly bokeh caused by transverse or lateral CA elsewhere. It’s splitting hairs, I guess, as if you use a deep red filter, then only the red spectrum will hit the sensor - the effect will be the same with both axial and transverse CA.  However, the filter won’t cause focus shift.  I guess it might be more apparent if it is already there.

Axial or longitudinal CA does particularly affect the red spectrum across the entire frame, and will also be apparent with poorly corrected lenses.

I agree with Jaap that, with properly corrected lenses, the red filter doesn’t have this apparent effect, but I guess you need to read the whole thread to get that.  Shorter focal lengths more sensitive to transverse CA; longer focal lengths to axial CA, but we’ve known that for a long time.

Short answer, nitpicking though it may be, is a red filter will cause no focus shift.  What they may do is make aberrations which already exist in a lens, particularly longer focal lengths, more apparent. 

I tend to use a yellow filter for skin tones and orange or red for landscapes with my Monochrom.  I don’t use blue or green, but I do use a polariser and ND filters.  75mm is my most commonly used lens for landscapes, though wides can be good where there is a subject in the foreground.  Obviously, others may take a different approach.
 

Edited July 4, 2023 by IkarusJohn

[Jeff S]

15 minutes ago, IkarusJohn said:

 Short answer, nitpicking though it may be, is a red filter will cause no focus shift.  What they may do is make aberrations which already exist in a lens, particularly longer focal lengths, more apparent. 

 

Yup, the point of my post #26 above, and the prior discussion.  The filter is not  the cause of the shift; the lens is… via CA.

Jeff

[jaapv]

The point is that red/green CA and colour filter focus shift are exactly the same phenomenon with a common root cause:chromatic aberration. Your camera is an RGB device. If the lens has CA it will focus the three channels in different planes. An OOF image will be larger ( if behind the middle plane) or smaller ( if in front). Thus coloured edges  will be produced. But if you use a coloured filter on a monochrome sensor or film your rangefinder will still focus on the middle focal plane, but your lens will focus on the (f.I) red projected image which is focused outside the sensor surface. Obviously the strength of the effect will depend on the correction of the lens. 
Note that the angle of view and thus the effective focal length depends on the focusing extension of the lens. Thus the size of the image in the focal plane will vary depending on the focus distance. The focal length of a lens is always given at infinity. I love the German word for this effect: Bildfeldschwund. 
So lateral chromatic aberration is caused by a combination of an insufficiently chroma-corrected lens and Bildfeldschwund. 

[costa43]

Admittedly I’m no expert but along with colour fringing, surely CA affects the sharpness of an image’s contrasting edges. 

[pegelli]

On 7/4/2023 at 10:37 PM, costa43 said:

Admittedly I’m no expert but along with colour fringing, surely CA affects the sharpness of an image’s contrasting edges. 

Indeed, and there is software that can correct CA based on the colour fringes it sees around edges and make the photo sharper. And I guess that such software will probably improve as AI gets better. However on a monochrome sensor the colour fringing is all grey so I think it will be harder to correct and therefore I think better corrected lenses for CA are more important for monochrom cameras as for colour cameras.

 

[Chaemono]

The ΑPO 50 Summicron and the 50 Summilux ASPH II blur the background differently both at f/2. Depending on the scene, the Summilux f/2 blur may work better. 

[jaapv]

On 7/4/2023 at 11:37 PM, costa43 said:

Admittedly I’m no expert but along with colour fringing, surely CA affects the sharpness of an image’s contrasting edges. 

Certainly. But the CA slider in your PP program will go a long way in correcting on colour images.